Social Network Feed Ranking via Viral Utility Computation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Social networking services face challenges in optimizing content item placement in user feeds to enhance user engagement and content circulation, as chronological ordering may not effectively promote viral actions or interactions with highly connected users.
Innovation Solution
A method that computes a utility value for viral actions based on user characteristics, content item characteristics, and first-degree connections, adjusting the position of content items in the feed to prioritize those with higher utility values, thereby incentivizing viral actions and maximizing downstream impact.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If content items are ordered chronologically in the feed, then the feed structure remains simple and easy to implement, but content circulation and viral actions are not effectively promoted
Solution Approach 1:
The patent changes the ordering parameter from purely chronological to a composite ranking system that incorporates multiple factors including viral action probability, user connection degrees, and engagement metrics. This allows the feed to promote content likely to go viral while maintaining a structured organization method.
Solution Approach 2:
The system performs preliminary computation of viral action probabilities and user connection analyses before generating the feed. By pre-calculating these metrics and ranking content items accordingly, the system prepares the optimal feed structure in advance, balancing complexity management with content circulation enhancement.
2Productivity
If the feed prioritizes content from highly connected users, then content circulation increases, but users with fewer connections have reduced visibility and engagement
Solution Approach 1:
The patent applies different weighting factors to different user segments based on their connection degrees. Highly connected users receive higher weights for content visibility, while users with fewer connections receive compensatory boosts through other metrics like content quality or novelty. This localized quality adjustment ensures both groups can achieve visibility.
Solution Approach 2:
The system dynamically adjusts ranking parameters based on user characteristics. For users with fewer connections, the system modifies the weighting of various factors (such as increasing the weight of content originality or engagement quality) to compensate for lower network influence, ensuring inclusive engagement opportunities.
3Productivity
If the system computes utility values based on multiple factors (user characteristics, content characteristics, connections), then content placement optimization improves, but computational complexity and processing time increase
Solution Approach 1:
The patent segments the utility computation into distinct modular components: user characteristic analysis, content characteristic analysis, connection degree calculation, and viral action probability estimation. Each module independently computes its aspect and passes results to the next, reducing overall computational complexity through divide-and-conquer.
Solution Approach 2:
The system pre-computes and caches user profiles, content metadata, and connection graphs before feed generation. By preparing these data structures in advance and storing them for quick retrieval, the system reduces real-time computational burden during feed assembly, maintaining optimization quality while reducing processing time.
Data Source
AI summary
Aspects of the present disclosure relate to increasing content circulation in a social networking service. A server accesses a content item for a user. The server computes a utility value for a viral action on the content item by the user, the utility value being computed based on user characteristics of the user, content item characteristics of the content item, and information about first degree connections of the user. The server adjusts, based on the computed utility value for the viral action, a position of the content item in a feed for presentation to the user. The server provides for presentation of the feed, including the content item, to the user.